Biomarkers are incorporated into cross-sectional, case-control and prospective cohort studies of occupational causes of cancer in order to enhance exposure assessment, provide insight into early biologic effects of specific chemicals, evaluate sources of genetic susceptibility and classify tumors at the molecular level to identify subgroups that may be more etiologically homogenous. A cross-sectional study of healthy workers exposed to benzene in Shanghai, China showed that exposed workers had elevated levels of chromosomal aberrations in peripheral lymphocytes, similar to the type observed in patients with chemically associated leukemias. A case-control study of benzene hematotoxicity in Shanghai showed that subjects with rapid CYP2E1 activity estimated by the drug chlorzoxazone (which activates benzene), who were homozygous for the 609C->T mutation in the NQO1 gene (which is involved in the detoxification of benzene metabolites), were at elevated risk for benzene hematotoxicity. A study of workers exposed to benzidine in India found that an acidic urine pH was associated with having elevated levels of benzidine-DNA adducts in exfoliated urothelial cells. A case-control study of non-Hodgkin's lymphoma (NHL) nested within the Washington county (Maryland) prospective cohort study found a significant association between pre-diagnostic levels of serum polychlorinated biphenyl compounds (PCBs) and risk of NHL. Newly developed and on-going case-control studies of stomach, esophagus, brain, bladder and breast cancer, NHL, and benzene-induced hematotoxicity and hematologic malignancies and cohort studies of women in China and agricultural workers in the United States are evaluating a range of potential genetic risk factors and their interaction with occupational and environmental exposures. Most of these studies are also collecting tumor samples for future molecular analyses.